<p> 1</p><p>Random Signals and Noise ELEN E4815</p><p>Columbia University Spring Semester- 2007</p><p>Prof I. Kalet 7 March 2007 Midterm Examination</p><p> Length of Examination- Two Hours  Answer All Questions</p><p>Good Luck and Have a Nice Vacation!!! 2</p><p>Problem #1 (34 Points)</p><p>The power spectral density, Px(f), of a random process, x(t), is shown below.</p><p>P x(f) A  (f)</p><p>C  (f+W) C  (f-W) B</p><p>–2W -W 0 W 2W f</p><p> a) What is the total average power in this random process? b) Is there a d-c component in this random process? Explain your answer. c) If your answer to part (b) was positive, how much power is there in the d-c component of the random process, x(t)? d) Is there a periodic component in this random process? Explain your answer.</p><p> e) Find and draw the autocorrelation function, x(), of this random process. 3</p><p> f) Now suppose that the autocorrelation function, x(), of a random process, x(t) is given below.</p><p>x() C  ()</p><p>A</p><p>-T 0 T </p><p>What is the total average power in this random process? g) Is there a d-c component in this random process? Explain your answer. 4</p><p>Problem #2 (33 points)</p><p>Consider the random process, x(t), the so-called random telegraph signal (shown below). In this signal, which started at - and will continue to +, the voltage flips back and forth, between +A volts and -A/2 volts, in the following manner.</p><p>The switching times, are dictated by a Poisson distribution, i.e., the probability of “k” switches in  seconds is given by the Poisson distribution function shown below.</p><p>Prob {of “k” switches in  seconds }= [e- {}k ] / k! where “k” = 0, 1, 2, 3, …... x(t) A</p><p> t - A/2</p><p>This is “a switching instant” 5 a) Find and draw the autocorrelation function of this random process-THINK!!! b) Find and draw the power spectral density of the random process described above. c) What is the total average power of this random process? d) If there is a DC component in this process, what is its average power? 6</p><p>Problem #3 (33 points)</p><p>You are given the following random process, x(t), defined by the equation below</p><p> x(t)= A cos (2Wt +) - < t <  </p><p>A is an independent random variable with a uniform probability density function between 0 and 1, and</p><p> is an independent random variable with a uniform probability density function between - and +. a) Is x(t) a wide-sense stationary process? Explain your answer!! b) If your answer to part (a) is positive, is x(t) ergodic in the mean and is it ergodic in the autocorrelation function? Explain your answer. c) A random process, x(t), is defined below. x(t)= A cos 2Wt for - < t < ; a priori probability=1/3 or x(t)= A sin 2Wt for - < t < ; a priori probability=1/3 or x(t)= 0 for - < t < ; a priori probability=1/3</p><p>Find the E{x(t)}. Find the E{x(t+) x(t)}. 7</p><p>Is this process WSS? Explain your answer.</p><p>If the process is WSS: Is it ergodic in the mean? Explain your answer Is it ergodic in the autocorrelation function? Explain your answer!</p><p>The End!!!</p><p>Some Helpful Equations cos x cos y= ½[ cos (x-y) + cos(x+y)] sin x sin y= ½[ cos (x-y) - cos(x+y)]</p>